incoming-request-advisor

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly requires pulling the full incoming message and instructs the agent to include direct quotes verbatim, which would force the model to reproduce any secrets present in the message.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly ingests the “incoming message itself” (e.g., Slack/email/thread text or attached screenshot/PDF) and extracts the full message before analysis, which is outsider-authored free text when it comes from someone other than the operating user.